The objective of this project is to develop a commercial software product to improve the visualization and resolution of two-dimensional (2D) images obtained from optical microscopes. Diffraction and other optical distortions present in every 2D microscope image are characterized by the point-spread-function (PSF) of the system, but this information is rarely known accurately. Iterative blind deconvolution is a method that enables the PSF to be estimated directly from the blurred and noisy image. The system developed can derive all necessary information from the image itself, and requires no additional parameters about the microscope setup, as is currently the case for 3D deconvolution. The algorithm developed is automated, can reconstruct spatial frequency components beyond the diffraction limit, and is robust to noise contamination. Intensity quantification is retained for certain classes of images. Restoration of low-light level live-cell time-lapse imagery is an important application area. In confocal microscopy, signal intensity can be increased at the expense of a loss in spatial resolution, which can be compensated for using the 2D blind deconvolution. This then enables light exposure to be reduced, thus keeping cells healthier for longer. Other application areas include widefield fluorescence, differential interference contrast, total internal reflection fluorescence, and spinning disk confocal microscopy. Restoration is fast enough to be carried out online. The Phase II objectives are to commercialize the current algorithm developed, improve the automation, increase the robustness against noise, and further reduce processing times. The potential market is significantly larger than that for current 3D deconvolution software. Imagery from almost every microscope "could benefit because no specialized equipment is required and a diffraction limit is always present.